Upregulation of Oct-4 isoforms in pulmonary artery smooth muscle cells from patients with pulmonary arterial hypertension

Abstract

Oct-4 is a transcription factor considered to be one of the defining pluripotency markers in embryonic stem cells. Its expression has also been demonstrated in adult stem cells, tumorigenic cells, and, most recently and controversially, in somatic cells. Oct-4 pseudogenes also contribute to carcinogenesis. Oct-4 may be involved in the excessive proliferation of pulmonary arterial smooth muscle cells (PASMC) in patients with idiopathic pulmonary arterial hypertension (IPAH), contributing to the pathogenesis of IPAH. In this study, we show that Oct-4 isoforms are upregulated in IPAH-PASMC. Human embryonic stem cells (H9 line) and human PASMC from normotensive subjects were used throughout the investigation as positive and negative controls. In addition to significant upregulation of Oct-4 in a population of IPAH-PASMC, HIF-2α, a hypoxia-inducible transcription factor that has been shown to bind to the Oct-4 promoter and induces its expression and transcriptional activity, was also increased. Interestingly, a substantial upregulation of Oct-4 isoforms and HIF-2α was also observed in normal PASMC exposed to chronic hypoxia. In conclusion, the data suggest that both Oct-4 isoforms are upregulated and potentially have a significant role in the development of vascular abnormalities associated with the pathogenesis of IPAH and in pulmonary hypertension triggered by chronic hypoxia.

pulmonary arterial hypertension (PAH) is a progressive and fatal disease characterized by elevation in pulmonary arterial pressure and pulmonary vascular resistance (PVR). Pulmonary vascular remodeling constitutes the major cause for the increased PVR, which can eventually lead to right heart failure and premature death (12). Idiopathic PAH (IPAH) is a rare form of the PAH with unidentifiable etiology and no associated underlying disease. Several mechanisms may contribute to the pulmonary vascular remodeling observed in IPAH patients, with one of the common pathologies being a significantly thickened arterial wall due to excessive proliferation and inhibited apoptosis of pulmonary artery smooth muscle cells (PASMC; Ref. 16). Studies have indicated that abnormalities in PASMC and pulmonary artery endothelial cells (PAEC) are present in patients with IPAH (21, 43), and multiple genes and pathways are involved in the development of pulmonary vascular remodeling in IPAH patients (20, 25, 28).

PASMC proliferation is regulated by multiple transcription factors (2, 14, 23, 37). Interestingly, a dedifferentiation or transdifferentiation of a mature contractile phenotype of PASMC to a highly proliferative phenotype has been implicated in pulmonary vascular medial hypertrophy in animals with hypoxia-induced pulmonary hypertension (29). There are several perspectives on the cell type contributing to the pulmonary vascular remodeling. An expansion of smooth muscle (SM)-like cells expressing smooth muscle α-actin (SM-αA) is observed in remodeled pulmonary artery walls; these are thought to be highly proliferative PASMC (8). It is also suggested that dedifferentiation of PASMC or a differentiation of pulmonary adventitial fibroblasts to a proliferative myofibroblast cell phenotype may occur (29). Furthermore, progenitor cells in the circulation, such as endothelial progenitor cells, mesenchymal stem cells (MSC), and resident vascular progenitor cells, are attracted to injured pulmonary vascular walls and adopt a SM-like phenotype (9, 19). Endothelial cells in the pulmonary vascular intimal layer are additionally capable of undergoing an endothelial-mesenchymal transition where the endothelial cells assume a SM-like phenotype (10). One of the most recent perspectives on the formation and persistence of plexiform lesions or “complex vascular lesions” postulates that PAH may be more accurately described as a neoplastic, angioproliferative disorder (26). It should be noted that the current study focuses on investigating the mechanisms of pulmonary vascular medial hypertrophy in patients with IPAH and not specifically on the formation of plexiform lesions.

In light of recent investigations showing functional Oct-4 expression in mature cell types (46), this study was designed to evaluate Oct-4 expression in highly proliferative PASMC isolated from patients with IPAH. Furthermore, a direct regulation of Oct-4 by hypoxia-inducible factor-2α (HIF-2α) has been shown (4), so it may be postulated that in hypoxic disease states, such as hypoxia-induced pulmonary hypertension, Oct-4 expression is enhanced.

Oct-4 is a transcription factor, officially termed POU5F1 and also known as OCT3, OCT3/4, OTF3, and OTF4. It is currently considered to be a critical factor in maintaining human embryonic stem cells (hESC) in their undifferentiated state (1, 3). Most recently, it has been shown to be an essential transcription factor required for the generation of induced pluripotent stem cells from differentiated mature cell types, such as skin fibroblasts (35, 42). Oct-4 expression is tightly controlled throughout embryogenesis and postnatal life; in the adult, Oct-4 is generally believed to be expressed exclusively in germ cells. Recent evidence suggests that Oct-4 is also present in other primitive cell populations, such as multipotent adult progenitor cells and MSC (11), and in more differentiated cell types, such as human peripheral blood mononuclear cells (46). Interestingly, ectopic Oct-4 expression has been demonstrated to contribute to tumor growth and thus is also now recognized as having oncogenic properties (6, 33, 36).

This study was designed to investigate whether cells from patients have an upregulation of Oct-4 isoforms. Oct-4 may thus be involved in the excessive proliferation of PASMC from IPAH patients, as shown in previous studies in our laboratory (44), and could potentially contribute to the pathogenesis of IPAH. Furthermore, we also aimed to study whether Oct-4 may be modulated by the hypoxic environment endured in the pulmonary artery in the development of hypoxia-induced pulmonary hypertension. The results from this study support a significant upregulation of Oct-4 isoforms in both PASMC from IPAH patients and PASMC maintained in a hypoxic environment.

EXPERIMENTAL PROCEDURES

Cell preparation and culture.

PASMCs were isolated from patients with IPAH as previously described (45); the cells were isolated from arteries of size at the range of 100–500 μm in diameter. Approval to use the human lung tissues and cells was granted by the University of California, San Diego (UCSD) Institutional Review Board. Control PASMCs were purchased from Lonza and maintained in culture at low passage (<6) in SM growth medium (SmGM). Data were obtained from a minimum of 2 different IPAH patient-derived cell lines. The mean pulmonary arterial pressure of the IPAH patients was 52 mmHg. hESC [National Institutes of Health (NIH)-approved H9 cell line] were purchased from WiCell. Approval for the use of hESC was granted by the UCSD Institutional Review Board and Embryonic Stem Cell Research Oversight committee. Cells were maintained in feeder-free conditions on hESC-qualified Matrigel (BD Biosciences) using mouse embryonic fibroblast-conditioned WiCell media [DMEM/F-12, knockout serum replacement l-glutamine, 2-mercaptoethanol, nonessential amino acids, basic fibroblast growth factor (Invitrogen)]. Cells were routinely passaged using manual dissection. hESC were karyotypically normal (checked every 10 passages; Cell Line Genetics) and maintained expression of 5 pluripotency markers [Oct-4, Nanog, Krupple-like factor 4 (Klf4), Sox-2, and c-Myc].

RT-PCR.

Total RNA was isolated from normal PASMCs, IPAH-PASMCs, and hESCs using TRIzol (Invitrogen), purified, treated with DNase (Invitrogen), and reverse-transcribed to cDNA (SuperScript II RT; Invitrogen). PCR was performed using Platinum Blue PCR SuperMix (Invitrogen), and thermal cycling conditions were as follows: 2 min at 95°C followed by 35 cycles of 30 s at 95°C, 30 s at 55°C annealing temperature, and 20 s at 72°C. For the Mueller primers (Oct4-i and Oct4-ii), the conditions were as follows: 5 min at 95°C followed by 40 cycles of 60 s at 95°C, 45 s at 62°C annealing temperature, and 80 s at 72°C. Primer pairs were specifically to amplify Oct-4, Oct-4 isoforms and pseudogenes, HIF-2α and other pluripotency markers Klf4, Sox-2, Nanog, and c-Myc (Table 1; Invitrogen). GAPDH was used as an internal control to semiquantify the results. PCR products were run on 1.2–1.5% agarose gels and visualized by GelStar nucleic acid stain (Lonza). The data were quantified using ImageJ software (NIH).

Table 1.

Oligonucleotide sequences of the primers used for RT-PCR

Standard Names (acc. no.)*	Size, bp	Predicted Sense/Antisense	Location, nt	Gene (Chrom.)
Oct-4 isoforms
Oct-4 (A and B)†	365	5′-GTGAGAGGCAACCTGGAGAATT-3′	(B) 519–540/862–883	6p21.33
		5′-CCTCAGTTTGAATGCATGGGAG-3′	(A) 775–796/1,118–1,139
Oct-4 (pseudogene free)	647	5′-GAAGGTATTCAGCCAAAC-3′	(B) 326–343/957–972	6p21.33
		5′-CTTAATCCCAAAAACCTGG-3′	(A) 582–599/1,213–1,228
Oct-4i (A + PSG1) (Ref. 22)	247	5′-CGTGAAGCTGGAGAAGGAGAAGCTG-3′	(A) 417–441/639–664	6p21.33
		5′-CAAGGGCCGCAGCTTACACATGTTC-3′	(PSG1) 363–386/585–610	8q24
Oct-4ii (A, B + PSG1) (Ref. 22)	218	5′-GACAACAATGAAAATCTTCAGGAGA-3′	(A) 688–713/883–905	6p21.33
		5′-TTCTGGCGCCGGTTACAGAACCA-3′	(B) 432–457/627–649
			(PSG1) 634–659/836–848	8q24
Oct-4 full-length (A) (Ref. 22)	1,133	5′-TCATTTCACCAGGCCCC-3′	(A) 14–30/1,128–1,146	6p21.33
		5′-GCAGGCACCTCAGTTTGAA-3′
Oct-4 pseudogenes
Oct-4-PSG1	215	5′-GACAACAATGAAAATCTTCAGGAGA-3′	634–658	8q24.21
(NR_002304)		5′-TTCTGGCGCCGGTTACAGAACCA-3′	826–848
Oct-4-PSG5	365	5′-TTGCTGCAGAAGTGGGTGGAGGAAG-3′	205–230	10q21.3
(NG_006104)		5′-GTACCAAAATGGGAGCCTGGGGC-3′	548–570
Pluripotency genes
Nanog	372	5′-GTGAAGACCTGGTTCCAGAA-3′	631–650	12p13.31
(NM_024865)		5′-GTCACTGGCAGGAGAATTTG-3′	983–1002
Sox-2	443	5′-GCCTGGGCGCCGAGTGGA-3′	648–665	3q26.3-q27
(NM_003106)		5′-GGGCGAGCCGTTCATGTAGGTCTG-3′	1,064–1,090
c-Myc	177	5′-GACTCTGAGGAGGAACAAGAAG-3′	1,320–1,341
(NM_002467)		5′-GTAGTTGTGCTGATGTGTGGAG-3′	1,475–1,496	8q24
Klf4	376	5′-GTCGGACCACCTCGCCTTACACAT-3′	1,995–2,018
(NM_004235)		5′-GGTCTTCCCTCCCCCAACTCACG-3′	2,348–2,370	9q31
Hypoxia inducible
HIF-2α	317	5′-CTTAGTGTTGTGGACACTGCAG-3′	3,950–3,972	2p21-p16
(NM_001430.3)		5′-AGTGCACTGAGCTATGTGACTC-3′	4,245–4,266
Smooth muscle markers
SM-αA	308	5′-GCATCCATGAAACCACCTAC-3′	871–890	10q22-q24
(NM_001613.1)		5′-GAAGCATTTGCGGTGGACAA-3′	1,159–1,178
SM-MHC	179	5′-GGTTGGCGACTGTTCTTTCC-3′	2,509–2,528	2p16.3
(NM_152994.2)		5′-CTTCTGTATTTGCCTCACTGGTTA-3′	2,665–2,688
Internal control
GAPDH	243	5′-GACAACGAATTTGGCTACAGC-3′	1,045–1,065	12p13.31
(NM_002046)		5′-GATGGTACATGACAAGGTGC-3′	1,268–1,287

^*The accession numbers in GenBank for the sequence used in designing the primers.

^†Accession numbers for transcription factors Oct-4A and Oct-4B are NM_002701 and NM_203289, respectively. Chrom., chromosome; PSG, pseudogene; Klf4, Krupple-like factor 4; SM-αA, smooth muscle α-actin; MHC, myosin heavy chain.

Sequencing.

PCR products were extracted from the gel and purified using Qiagen QIAquick Gel Extraction Kit as per the manufacturer's guidelines. The purified products were sent to Eton Laboratories for primer extension sequencing (Eton Bioscience, San Diego, CA). Sequences were analyzed using BioEdit software and aligned to the NIH BLAST database sequences for Oct-4 isoforms (NM_002701, longer isoform A and NM_203289, shorter isoform B).

Immunofluorescence staining.

Cells were stained with mouse monoclonal anti-Oct-4 antibody (1:50; Santa Cruz Biotechnology, Santa Cruz, CA) and visualized by detection with anti-mouse IgG-rhodamine (1:100). Nuclei were counterstained with the membrane-permeable nucleic acid stain 4′,6′-diamidino-2-phenylindole (DAPI; 5 μM). Fluorescent images were taken on a DeltaVision RT Deconvolution Microscope. The antibody for Oct-4 from Santa Cruz Biotechnology targets amino acids 1–134, selecting for the unique NH₂ terminus of the Oct-4 isoform A gene, specifically located in the cell nucleus.

Hypoxia treatment.

Normal PASMCs were incubated in normoxia (21% O₂) and hypoxia (3% O₂) in a humidified atmosphere at 37°C for 72 h before the respective experiments were carried out.

Statistical analysis.

Linear regression equations were used to calculate correlation of differential gene expression between normal and IPAH-PASMC. For real-time RT-PCR experiments, mRNA expression levels between normal and IPAH-PASMC are expressed as means ± SE. Statistical analysis was performed using unpaired Student's t-test. Differences were considered to be significant when P < 0.05.

RESULTS

Upregulation of transcription factor Oct-4 in PASMC from IPAH patients.

Oct-4 expression was investigated in PASMC from control subjects (normotensive patients) and IPAH patients and in H9 hESC. Using RT-PCR and semiquantitative analysis by expressing the PCR product intensity as a ratio to internal control, GAPDH, the mRNA content of each cell population was determined. The mRNA expression of Oct-4 (isoforms A and/or B) was significantly higher in PASMCs isolated from IPAH patients, >4-fold higher compared with cells from normotensive controls (Fig. 1A, a and c). Oct-4 mRNA expression in hESC, which highly express this gene in the pluripotent state, was used as a positive control (Figs. 1Ab and 2). In PASMC from normotensive controls, Oct-4 mRNA expression was barely detectable. Immunofluorescence staining shows specific nuclear localization of Oct-4 (red), as illustrated in merged images where the nuclei are counterstained with nuclear specific dye DAPI (blue; Fig. 1B). Normal PASMC have low or no Oct-4 protein expression, whereas IPAH-PASMCs have a substantially increased Oct-4 staining (Fig. 1, B, C, and Da). The histogram in Fig. 1Db summarizes the distribution of Oct-4 fluorescence intensity in IPAH-PASMC, and the bar chart in Fig. 1Da compares the averaged intensity. Negative immunofluorescence controls (−) are shown to demonstrate the specificity of staining (Fig. 1C). The scale bars in the fluorescent images indicate 50 μm in the ×20 images and 20 μm in the ×40 images. IPAH-PASMC, like hESC, have nuclear localization of Oct-4 indicative of the Oct-4A isoform (Figs. 1B and 2); Oct-4B is located in the cytoplasm. Furthermore, the nuclear staining using an antibody specific to the first 134 nucleotides of the NH₂ terminus is suggestive of expression of Oct-4 isoform A as this region is only present on this particular isoform. Furthermore, HIF-2α, known to bind to the Oct-4 promoter and induces its expression and transcriptional activity, was also increased (4). Significantly increased HIF-2α protein expression was detected by Western blot in IPAH-PASMC compared with normal PASMC (Fig. 3). These results indicate that the mRNA and protein expression of Oct-4 in PASMC isolated from IPAH patients is significantly upregulated compared with normal PASMC, and the upregulated Oct-4 in IPAH-PASMC is associated with increased protein expression of HIF-2α.

Fig. 1.

Increased expression of transcription factor Oct-4 in pulmonary artery smooth muscle cells (PASMCs) from idiopathic pulmonary arterial hypertension (IPAH) patients. A: RT-PCR gel images showing PCR products of Oct-4 and GAPDH in normal (Nor) and IPAH-PASMC (a) and human embryonic stem cells (hESC; b). M, 100-bp DNA marker. Summarized data show expression of Oct-4 relative to the GAPDH intensity in each cell type (c). B: immunofluorescent staining of Oct-4 (red) in cells in normal PASMCs (a) and IPAH-PASMC (b). Cells are counterstained with nuclear dye 4′,6′-diamidino-2-phenylindole (DAPI; blue). Images are shown at ×20 (top) and ×40 (bottom) magnification, and representative cells in each condition are enlarged for clarity below the respective image. Bars, 20 μm in ×20 images and 10 μm in ×40 images. These measurements apply to all figure legends. C: negative control (−) is shown to demonstrate the specificity of staining. D: the averaged fluorescence intensity of Oct-4 staining of >90 individual cells (a) and the distribution of the fluorescent intensity compared in histograms (b). **P < 0.01, ***P < 0.001 vs. normal (Norm) PASMC. Ab, antibody.

Fig. 2.

Pluripotent hESC have nuclear Oct-4 expression. A: immunofluorescence staining of Oct-4 (red) in H9 hESC colonies. Cells are counterstained with nuclear dye DAPI (blue). Images are shown at ×20. B: RT-PCR gel images showing PCR products of pluripotency markers of hESC. GAPDH was used as an internal control. Klf4, Krupple-like factor 4.

Fig. 3.

Increased expression of hypoxia-inducible factor-2α (HIF-2α) in IPAH-PASMC. A: representative Western blot shows HIF-2α protein expression in normal PASMC (Nor) and in IPAH-PASMC (IPAH). B: summarized HIF-2α expression in arbitrary units (a.u.) normalized to the GAPDH internal control for 14 normal and 10 IPAH independent samples. P < 0.05.

Hypoxia upregulates Oct-4 and HIF-2α expression in normal PASMC.

Hypoxia-induced pulmonary hypertension is another pulmonary vascular disease characterized by sustained pulmonary vasoconstriction and further narrowing of the pulmonary arteries due to pulmonary vascular remodeling. To assess whether a similar upregulation of Oct-4 occurs in response to hypoxia, normal PASMC were incubated for 72 h in 3% O₂, 37°C humidified atmosphere. mRNA and protein expression were assessed. GAPDH, Oct-4, and HIF-2α were amplified from normal PASMCs treated under normoxia and hypoxia by RT-PCR (Fig. 4A). Semiquantification of PCR products by normalization to the internal control, GAPDH, indicated that the Oct-4 mRNA expression level in hypoxia-treated cells increased 5.7-fold compared with normoxia-treated cells (P < 0.01; Fig. 4B). Immunocytochemistry, using the Oct-4 H-134 antibody (Santa Cruz Biotechnology), confirmed the observations at the protein level. Nuclear expression of the Oct-4 protein was significantly (P < 0.001) higher after exposure to hypoxia (Fig. 4, C–E). Like IPAH-PASMC, the HIF-2α mRNA level was significantly upregulated in normal PASMC after hypoxia treatment (P < 0.01; Fig. 4, A and B).

Fig. 4.

Chronic hypoxia significantly upregulates Oct-4 expression in normal PASMC. A: representative RT-PCR gel images showing expression of OCT-4 and HIF-2α in normoxia (Nor) and hypoxia (Hyp). B: summarized data (means ± SE; right) showing PCR products of Oct-4, HIF-2α, and GAPDH in PASMC exposed to normoxia (20% O₂, Po₂ = 145 mmHg) and hypoxia (3% O₂, Po₂ = 22 mmHg) for 72 h. C: immunofluorescence staining showing Oct-4 protein level in normal PASMCs under normoxic and hypoxic conditions; the highlighted squares show selected cells magnified. D: negative control (−) is shown to demonstrate the specificity of staining. E: mean Oct-4 nuclear fluorescent intensity of cells in normoxia and hypoxia. **P < 0.01, ***P < 0.001 vs. hypoxia.

Upregulation of both Oct-4 isoforms in PASMC from IPAH patients.

Although a mouse monoclonal antibody (sc-5279; Santa Cruz Biotechnology) that recognizes amino acids 1–134 of Oct-4A protein and has been shown not to bind to the Oct-4B isoforms was chosen for these experiments, it cannot be entirely ruled out that Oct-4B was not detected due to the 95% sequence homology in this binding region. The next set of experiments was performed to specifically identify the various Oct-4 isoforms and pseudogenes through design of specific primer pairs to select for the different isoforms. We used our own primer pairs designed to eliminate pseudogene detection as well as the primers from Mueller et al. (22) and Suo et al. (Ref. 34; Fig. 5A). RT-PCR results using our own primers and those of Mueller et al. (22) gave PCR products in both hESC and in the IPAH-PASMC. Our original primers designed to target Oct-4 gave strong single bands in hESC and double bands in IPAH-PASMC, the expected band at 360 bp, and an additional band near 700 bp (Fig. 5B). These primers potentially amplify both isoforms and pseudogenes. We designed primers to be pseudogene-free; these produced strong bands in hESC and detectable bands in several of the IPAH-PASMC lines at lower passages, and no bands were detected in normal PASMC or higher (>8 passages) IPAH-PASMC (Fig. 5C). The primers used by Mueller and colleagues consisted of 1) Oct-4i, designed to amplify a sequence from exon 1 to 3 and determined to be specific for Oct-4A; 2) Oct-4ii, designed to amplify a sequence from exon 3 to exon 5 and to detect Oct-4 A and B isoforms; and 3) Oct-4 full-length, designed to yield the full cDNA of Oct-4 (Fig. 5A). Using these primers in our experiments, Oct-4i produced PCR products in half of the tested IPAH-PASMC cDNAs, whereas the Oct-4ii was detected in all but the high passage cells. Both produced strong bands in hESC. The full-length primer set only produced detectable products in hESC and none in IPAH-PASMC (Fig. 5D). Although this may indicate a lack of or low expression of the full gene, it may also suggest that the smaller population of cells expressing Oct-4 in IPAH-PASMC had insufficient transcript to successfully amplify by PCR. The sequencing of the PCR products confirmed a 99–100% homology to Oct-4 A and B isoforms. Only 92–96% homology was also observed for pseudogene 1 (PSG1). Sequences were obtained from PCR products from Oct-4 (365 bp; our laboratory), Oct-4 pseudogene-free primer pairs (647 bp; our laboratory), and Oct-4ii (Ref. 22; data not shown).

Fig. 5.

Specific detection of Oct-4 parental isoforms in IPAH-PASMC. A: schematic diagram showing the location of the sequences amplified by the primer sets in the Oct-4 gene. B: representative images of PCR products using standard Oct-4 primers (365 bp) and GAPDH (243 bp). C: representative images of PCR products using standard Oct-4 pseudogene-free primers (PSF; 647 bp) and GAPDH (243 bp). D: representative images of PCR products using Mueller et al. (22) Oct-4 primers, Oct-4i (247 bp), Oct-4ii (218 bp), and Oct-4 full-length (1133 bp). Samples 3 and 4 are higher than passage 8 IPAH-PASMC.

Pseudogene mRNA is downregulated in IPAH-PASMC.

Using the primers that Suo et al. (34) designed to detect the specific expression of Oct-4, PSG1 and PSG5 was assessed and compared with primers designed to eliminate pseudogene contamination (Fig. 6). It has been shown that these pseudogenes are transcribed in cancers and that they may play a role in regulating gene activity and carcinogenesis. We found that PSG1 and PSG5 were expressed in normal PASMC and significantly decreased in IPAH-PASMC (P < 0.01; Fig. 7). hESC also expressed these pseudogenes. In concert, Oct-4 was barely detected in normal PASMC, but expression was significantly increased in IPAH-PASMC (P < 0.001; Fig. 7), supporting an upregulation of Oct-4 A and/or B isoforms.

Fig. 6.

Targeting of the Oct4 gene for RT-PCR and sequencing. Aligned sequences for Oct-4A (black) and Oct-4B (red) genes with primer pair alignment indicated. The right box shows representative sequenced RT-PCR products for the primer pairing Oct-4ii for 2 IPAH patients and 1 hESC line. PSG, pseudogene. P1 and P2, patients 1 and 2.

Fig. 7.

Detection of Oct-4 pseudogenes in PASMC from IPAH patients. A: representative RT-PCR images comparing Oct-4 pseudogene-free expression in hESC and normal PASMC. B: representative RT-PCR gel images showing PCR products of Oct-4 pseudogene-free and Oct-4 pseudogenes 1 and 5 (PSG1 and PSG5) in PASMC from normal subjects, IPAH patients, and hESC (as a positive control). C: mean Oct-4 and pseudogene expression normalized to GAPDH (an internal control). **P < 0.01, ***P < 0.001 IPAH-PASMC vs. normal PASMC. ES cells, embryonic stem cells.

Expression of other pluripotency genes in PASMC.

The presence of other genes deemed pluripotency factors due to their role in maintaining the pluripotency of hESC and their ability to successfully reprogram differentiated cells to the pluripotent state was also investigated in the IPAH and normal PASMC (35). The mRNA expression of mature SM cell markers, SM-αA and SM myosin heavy chain, along with pluripotency factors, Klf4, Nanog, Sox-2, and c-Myc, was assessed (Fig. 8). Whereas hESC strongly expressed all the pluripotency factors (Fig. 2), expression of Sox-2 and Nanog was absent in IPAH and normal PASMC (Fig. 8). Interestingly, both normal and IPAH-PASMC cells expressed, at comparable levels, Klf4 and c-Myc, in addition to the SM markers (Fig. 8; Ref. 8).

Fig. 8.

PASMC do not express pluripotency markers Sox-2 and Nanog. A: representative RT-PCR gel images showing PCR products for smooth muscle α-actin (SM-αA; SM-α-actin; SMA). Negative control (−) is shown to demonstrate a lack of genomic contamination. B: representative RT-PCR gel images showing PCR products for pluripotency markers. C: averaged semiquantitative RT-PCR in normal and IPAH-PASMC; data are normalized to the GAPDH internal control. MHC, myosin heavy chain.

DISCUSSION

Recent studies suggest that Oct-4 does not only function to maintain pluripotency and self-renewal in hESC, but also is expressed in adult cell types (6, 13). Our data show a significant upregulation of Oct-4 in cells isolated from patients with IPAH and in human PASMCs subjected to hypoxic environments. In-depth investigation indicates that IPAH cells have upregulated expression of both Oct-4A and Oct-4B isoforms. The expression of Oct-4 in somatic cells has been challenged, primarily due to the existence of two Oct-4 isoforms, A and B. These result from alternate splicing of the Oct-4 gene producing two Oct-4 protein isoforms that differ in their NH₂ termini. The POU-binding domains remain the same, and COOH-transactivation domains are long in Oct-4 isoform A and short in isoform B (13, 15). Detection of Oct-4 expression by RT-PCR is prone to artifacts generated by pseudogene transcripts; design of specific primers can avoid such misinterpretation of data (17, 18, 22).

Using sets of primers designed to specifically detect Oct-4A, Oct-4B, and pseudogenes of Oct-4, mRNA expression of both isoforms and pseudogenes was detected. At this stage, it may be concluded for certain that Oct-4 and pseudogenes are present in IPAH-PASMC, however, control PASMC only have detectable pseudogene mRNA. Furthermore, in IPAH-PASMC, a concomitant augmented expression of Oct-4 and decreased expression of pseudogene mRNA supports a specific enhanced expression of the Oct4 gene. Sequencing of the PCR products confirmed Oct-4 A and/or B expression but failed to discriminate between the isoforms (99–100% homology to both); sequence homology to the pseudogenes was lower (92–96%). Two other pluripotency genes were detected in PASMC, Klf4 and c-Myc. Although not previously studied in the pulmonary artery, the expression of Klf4 has been shown in SM cells being associated with vascular injury and regulation of SM differentiation markers and proliferation in vivo (40). c-Myc is a known oncogene associated with proliferative, antiapoptotic pathways; its expression and function has been demonstrated in human PASMC (30).

A strong nuclear localization of the Oct-4A-specific antibody was observed in a population of IPAH-PASMC, reflecting that observed in hESC and indicative of an augmented Oct-4A expression in both IPAH-PASMC and hypoxic PASMC. Mueller and colleagues (22) showed a correlation between a primer set detecting the full-length Oct-4 cDNA and protein expression. Although no full-length cDNA was detected from our PCR reaction in IPAH-PASMC, it is likely that a lower copy number exists in IPAH-PASMC compared with hESC, which abundantly express Oct-4 in their pluripotent state and thus making amplification of a large cDNA technically demanding.

The enhanced expression of Oct-4 isoforms in both IPAH-PASMC and in hypoxia-treated PASMC is interesting and suggestive of a functional role for Oct-4 in the development and progression of pulmonary vascular remodeling in patients with IPAH and PH associated with hypoxia. In the pulmonary circulation, hypoxia causes pulmonary vasoconstriction and vascular remodeling (or concentric pulmonary vascular hypertrophy) and is also associated with the upregulation of hypoxia-inducible factors (27, 41); it is plausible that there may be associated changes in Oct-4 expression and potentially with a dedifferentiation of cells in the pulmonary artery to a characteristic hyperproliferative phenotype. The increased Oct-4 expression in hypoxia-treated cells and in IPAH-PASMC correlated to an upregulation of HIF-2α. As a HIF-2α-specific target gene, the upregulation of Oct-4 may result from the increased expression of HIF-2α under hypoxic conditions. Delayed onset of differentiation or spontaneous differentiation of hESC is observed when the cultures are maintained in a low O₂ (<5% O₂) environment (38), and it is known that a reduction in oxygen concentration stabilizes HIF-1α and -2α; of these, HIF-2α is known to directly to regulate Oct-4 expression (4). Furthermore, HIF-2α has been shown to have a predominant role in the hypoxic cellular response of human pulmonary artery adventitial fibroblasts in the processes of proliferation and migration (7). It is known that in both IPAH and hypoxia, PASMC contribute to vascular remodeling by 1) proliferating in the SM layer, resulting in medial layer thickening (24, 39); 2) migrating to other vascular layers: intimal and adventitial layers and to precapillaries and capillary arterioles causing muscularization; and/or 3) transdifferentiating to other cell types after migration: myofibroblasts or endothelial cells (5, 31, 32).

Our results show that Oct-4 expression in elevated IPAH-PASMC and PASMC in hypoxic environments, corroborating other studies indicating elevated Oct-4 expression in adult/somatic cells. Using multiple techniques, a decreased pseudogene expression supports an increased Oct-4 expression due to either or both the Oct-4A and/or Oct-4B isoform. The precise functional roles of Oct-4A and/or Oct-4B in IPAH and hypoxia-induced pulmonary hypertension are potentially very interesting. Its specific upregulation in diseased (IPAH) and hypoxic PASMC make it a potential therapeutic target in the treatment of IPAH and hypoxia-induced pulmonary hypertension.

GRANTS

We thank the UCSD Neuroscience Microscopy Shared Facility (National Institute of Neurological Disorders and Stroke Grant P30-NS-047101). This work was supported, in part, by the NIH National Heart, Lung, and Blood Institute Grants HL-64945 and HL-66012. A. L. Firth is supported by a California Institute for Regenerative Medicine Postdoctoral Training Fellowship.

DISCLOSURES

No conflicts of interest are declared by the author(s).